Daniela Witten @daniela_witten Twitter profile | Pikagi

Pikagi

Daniela Witten

@daniela_witten

48,768

Followers

754

Following

229

Media

4,150

Statuses

dorothy gilford endowed chair and prof of stat/biostat @uw . all views my own.

Seattle, WA

https://t.co/sdJDhEbuF3

Joined July 2017

Don't wanna be here? Send us removal request.

Pinned Tweet

@daniela_witten

Daniela Witten

@daniela_witten

3 years

Finally --- the 2nd edition of ISLR is out!!!!🥳🎉🪅🍾 Well, sort of. 🙄It's "published", but due to shortages in the global paper supply, hard copies won't be available for a bit longer 😢😭🤦‍♀️ You can pre-order a hard copy from Springer/Amazon, & **download the pdf today**! 1/

Tweet media one

62

528

2K

Last Seen Profiles

@SarahCo97953037

@47Arty

@Deborah04880917

@NotCharles2

@KuayreNaiZ

@ANC0101

@MundellaSchool

@Kyle42426690851

@willduffy92

@Sadow51

@getoneword

@Ekkard_Baeuerle

@waetanabe

@Grace477000

@FNATIC

@Bernin1001

@TC3uk

@DanetteDoe41775

@Akadust_

@LanceMcccabe

@15MinuteHistory

@metz1969

@PsychNation2022

@shaykiafterdark

@jihoongffr

@Bbykitty16

@Lisa98958966823

@VishuBA320383

@Stephenson22448

@harmony451401

@mggirlssoccer

@TVET4Girls

@malcolm_fan22

@Nixon669930

@LaurieE80735449

@XXL_Wasjuk

@daniela_witten

Daniela Witten

@daniela_witten

2 years

When I tell a tech bro that I am a professor of statistics, they often immediately inform me that I am a teacher professor, not a researcher professor. I say no, I am a professor professor, I do research and teach, but mostly research. Then they tell me- oh but I’m an UNDERGRAD

150

508

11K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

FOUR CARDINAL RULES OF STATISTICS 📈📊📉 A thread 1/🧵

78

2K

9K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

hey babe, are you ok? someone with an MBA just referred to logistic regression as "machine learning" and you stayed calm

138

555

9K

@daniela_witten

Daniela Witten

@daniela_witten

3 years

This is everything

Tweet media one

79

924

8K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

Academia is funny because it's like we put a whole bunch of narcissists who are used to being #1 at everything into a room together, lock the door, throw away the key, and then say oH mY GoD i cAN't bELieVe ThIS iS How iT TUrnEd ouT

63

576

7K

@daniela_witten

Daniela Witten

@daniela_witten

3 years

Nobody: Absolutely nobody: Literally not one person: @naturecomms : What’s holding back women in science is women in science

23

628

6K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

Each PhD admissions season, the number of emails I get from aspiring PhD students that mention my physical appearance is not zero.... and I guess I’m not really sure what to say about this other than that it really seems like it should be zero.

122

313

5K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

professor, but not like the ones they had in undergrad, who taught grad students too. Sometimes I try to keep explaining to them, but sometimes I stop because I don’t want to explode their brains 🤯 Men- why are you like this!!

48

68

5K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

The Bias-Variance Trade-Off & "DOUBLE DESCENT" 🧵 Remember the bias-variance trade-off? It says that models perform well for an "intermediate level of flexibility". You've seen the picture of the U-shape test error curve. We try to hit the "sweet spot" of flexibility. 1/🧵

Tweet media one

59

1K

5K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

It actually makes sense that the physicists didn’t know about college parties

103

551

5K

@daniela_witten

Daniela Witten

@daniela_witten

5 years

"When we raise money it’s AI, when we hire it's machine learning, and when we do the work it's logistic regression." (I'm not sure who came up with this but it's a gem 💎)

40

942

4K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

This happened most recently three days ago, at a kiddo social event, where someone told me that MIT (where he went to school) is run really differently from UW because at MIT people only get the title “professor” if they do research

90

55

4K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

Just kidding!! It's super bad. ☹️ /end

44

86

3K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

So sorry to all who I offended with a joke about R versus python. They are both fine, it’s just personal preference! Like, do you prefer to do a quick and correct data analysis and then sit by the beach sipping mojitos, or

62

165

3K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

Like many people, I have spent the last 10 days watching so much tragedy unfold. So much anguish from Black colleagues here on twitter. And so I've been trying to think of ways that *I* can improve my tiny corner of the world. A thread on why change is hard in academia 1/

94

943

3K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

Twitter laid me off today. If you know of any open positions for project managers or senior software engineers, let me know. I was in charge of the feature that made sure every time you tweeted about machine learning you were served an ad for an intro stats course.

59

138

3K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

I just updated my Twitter bio to include my CV highlights, in the hope that this will reduce the number of cold e-mails I get asking what on earth makes me think I'm qualified to weigh in on important issues in the field of statistics etc. Also, people are exhausting.

64

84

3K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

She’s a 10 but she prefers R to python, so, yeah, she’s still a 10

26

178

3K

@daniela_witten

Daniela Witten

@daniela_witten

3 years

The recent @NatureComms paper doesn’t tell us much about the impact of gender on mentorship but it sure does tell us that the statistics community needs to do a better job teaching scientists about correlation, causation, and confounding

21

269

3K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

PERSPECTIVE FROM A STATISTICIAN: Why the Covid-19 🦠 situation in the U.S. 🇺🇸 is **not** actually that bad --- a look at the numbers!!! 📊📈📉 1/

37

559

3K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

Statistics means never having to say you’re certain.

49

349

3K

@daniela_witten

Daniela Witten

@daniela_witten

3 years

confused about why I am being charged $350 to give an invited talk at a virtual conference (!!!?!?)

118

58

2K

@daniela_witten

Daniela Witten

@daniela_witten

3 years

My science origin story: As a little girl, I always dreamed of spending literally two decades of my life rising to the top of my field of academic inquiry, only to have people who haven’t read my papers talk down to me about them on twitter

39

87

2K

@daniela_witten

Daniela Witten

@daniela_witten

3 years

The second edition of ISLR is due out by end of summer! If you are an **instructor** in need of a pre-print to prepare for AY 2021-2022, then send an e-mail to hello @statlearning .com

Tweet media one

24

262

2K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

academia is a pyramid scheme, but not the type where you get rich. more like the type where you end up with a ton of stickers

34

112

2K

@daniela_witten

Daniela Witten

@daniela_witten

1 year

Ugh I should know better than to wade into the fray here, but I would MUCH rather have 4 grad students who are well paid and happy instead of 5 grad students who are paid 20% less and need to scrape together summer internships and extra TA jobs to afford their rent 1/3

20

140

2K

@daniela_witten

Daniela Witten

@daniela_witten

5 years

Stitch Fix is using something called eigenvector decomposition, a concept from quantum mechanics or literally Week 5 of any introductory course in linear algebra

@WIRED

WIRED

5 years

Stitch Fix is using something called eigenvector decomposition, a concept from quantum mechanics, to tease apart the overlapping “notes” in an individual’s style. Using physics, the team can better understand the complexities of the clients’ style minds.

242

202

1K

29

195

2K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

I don’t know who needs to hear this, but: if you somehow developed the skills to do independent research before starting your PhD, then wow!! 💫 🦄 But I didn’t. So if you’re like me, then come to grad school, and we’ll train you to become an independent researcher. 👩‍🏫 👩‍🎓

29

123

2K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

do you prefer to bang your head against a brick wall for hours trying to figure out how to subset your data matrix in a language that LITERALLY IS NOT DESIGNED FOR DATA ANALYSIS??? Both options are equally valid! It’s a matter of personal preference!!!

46

34

2K

@daniela_witten

Daniela Witten

@daniela_witten

3 years

today I had to abruptly end a meeting with a collaborator because an arborist I'd hired for some tree work was at the door. before I signed off, my collaborator said "careful you don't end up with a random forest" and I'm still laughing.

18

50

2K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

When I started my PhD, 1 of my classmates had 1st author papers from undergrad in 2 of the top 5 journals in our field. I hadn’t even heard of those journals!!!! Today, he’s a tenured professor in a top-10 department, NSF CAREER awardee, etc. Guess what... I am too!!

@daniela_witten

Daniela Witten

@daniela_witten

4 years

I don’t know who needs to hear this, but: if you somehow developed the skills to do independent research before starting your PhD, then wow!! 💫 🦄 But I didn’t. So if you’re like me, then come to grad school, and we’ll train you to become an independent researcher. 👩‍🏫 👩‍🎓

29

123

2K

18

124

2K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

If you like #islr first edition then you will love 💗 🥰 our second edition... coming soon!!

@janaki_shibu

Janaki

4 years

This was honestly the best textbook I have read in a really, *really* long time (exception: de la Fuente's Mathematical Methods & Models for Economists). It is concise, well-written & comprehensive. It is lovely (& that's not a word I would generally use to describe Econ books!

Tweet media one

12

45

456

56

155

2K

@daniela_witten

Daniela Witten

@daniela_witten

5 years

Perfect gym for a statistician

Tweet media one

18

290

2K

@daniela_witten

Daniela Witten

@daniela_witten

3 years

There are lots of medical studies out there getting more attention than they should. All else being equal, we should trust the ones with larger sample sizes. Why? Because larger sample size leads to reduced variance. By how much? 1/n

62

189

2K

@daniela_witten

Daniela Witten

@daniela_witten

3 years

ML is literally just statistics

@GenericBologna

rhymes with purple

@GenericBologna

3 years

In 5 words or less, start a fight without politics

29K

757

8K

32

120

2K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

As a statistician, I have a lot of expertise that may prove useful in the current pandemic. For instance, I've been social distancing since I started grad school in 2005, and am happy to provide tips

8

122

2K

@daniela_witten

Daniela Witten

@daniela_witten

3 months

Told my kids that they are extremely lucky— people pay money to hear me lecture, whereas my kids get to experience my lectures constantly and for free. 8yo remains skeptical.

22

37

1K

@daniela_witten

Daniela Witten

@daniela_witten

3 years

Happy New Year!!! 🎆🍾🎉🥳 As this year draws to a close, I'm delighted to humbly announce that this year I published 135 papers, increased my h-index by 384, saved a species of small river otter from the brink of extinction, and lied at least 3 times.

36

50

1K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

Reviewer criticism can be unreasonable, unfair, and even unhinged, which is why an academic career is great training for cooking dinner for three small humans on a weeknight.

15

61

1K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

6 year old is about to play her first game of computer chess. She said “I’m probably going to win since computers don’t have brains.” Should I tell her?

28

46

1K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

Everyone! It was a joke!!! 🤦‍♀️ Whether you should use python or R *depends on the situation*. Like, you should use python to perform a side-by-side comparison of ease-of-use of R versus python for data analysis. After you’re done the comparison, use R

34

53

1K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

Why does this matter? In many contemporary settings (think: the internet), sample sizes are so huge that we can get TINY p-values even when the deviation from the null hypothesis is negligible. In other words, we can have STATISTICAL significance w/o PRACTICAL significance. 7/

11

109

1K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

ONE: CORRELATION DOES NOT IMPLY CAUSATION. Yes, I know you know this, but it’s so easy to forget! Yeah, YOU OVER THERE, you with the p-value of 0.0000001 — yes, YOU!! That’s not causation. 2/

10

106

1K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

TWO: A P-VALUE IS JUST A TEST OF SAMPLE SIZE. Read that again — I mean what I said! If your null hypothesis doesn’t hold (and null hypotheses never hold IRL) then the larger your sample size, the smaller your p-value will tend to be. 5/

19

134

1K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

This is dangerous misinformation because THAT’S 👏 NOT 👏 HOW 👏 CONDITIONAL 👏 PROBABILITY 👏 WORKS

@soonergrunt

soonergrunt 🇺🇸 A Capybara Appreciation Account

2 years

Picked up a hitchhiker today. He seemed like a nice guy. After a while though, he asked me if I wasn't afraid that he might be a serial killer. I told him that the odds that two unrelated serial killers were in the same car at the same time were astronomical.

780

9K

122K

41

79

1K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

I have a machine learning joke but actually it’s a statistics joke.

@MenonBioPhysics

Gautam Menon

@MenonBioPhysics

4 years

I have an epidemiology joke but it hasn't yet gone viral.

918

2K

13K

24

100

1K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

fauxductivity (noun). /ˌfōˌdəkˈtivədē/ The illusion of a productive day, when all you've actually accomplished is checking administrative tasks off of your to-do list (as opposed to your actual research)

21

103

1K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

P-values

@LauraSavino747

Laura, just a regular pilot turned writer

@LauraSavino747

2 years

Name something people have strong opinions on, but actually know little about.

4K

217

3K

11

90

1K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

Has anyone had any luck staying in shape, or even improving physical fitness, during COVID times? 🏃‍♀️💪 Post your reply in the comments so that I know who to unfollow

152

17

1K

@daniela_witten

Daniela Witten

@daniela_witten

2 years

before kids, I’d use 5 free minutes between meetings to get a coffee and check my texts. with kids, I use those 5 minutes to get coffee, check texts, respond to 9 emails, take a shower, call the pediatrician, do a 30 minute peloton workout, pay bills, visit the dentist,

14

35

1K

@daniela_witten

Daniela Witten

@daniela_witten

3 years

I’m 37 today and I’d love to share some hard-won wisdom but the best I can come up with is that you shouldn’t reply to a group text unless you know who all the numbers belong to

46

18

1K

@daniela_witten

Daniela Witten

@daniela_witten

4 years

I just got a cold e-mail from a stranger trying to prove some statistical point to me with an example in R. It included this sentence: "If you don't know how to use R then ask a colleague to help you." I think that's my cue to call it for the night. 🙃

29

26

1K

@daniela_witten

Daniela Witten

@daniela_witten

3 years

Even if you’re fully vaccinated, the CDC recommends that you never create a doodle poll with more than five time slots.

16

62

978

@daniela_witten

Daniela Witten

@daniela_witten

3 years

Personally I’m an #RStats user... But I think it’s important for the stats community to be inclusive of those who use python... (Or, as I like to call it, R Total Landscaping)

22

60

959

@daniela_witten

Daniela Witten

@daniela_witten

2 years

Delighted to be @stanford today as co-organizer (with J Bien and Ryan Tibs) of @robtibshirani ’s 65th bday party. 🎈 🎊 🎉 This morning’s big reveal: lasso-inspired matching outfits.

Tweet media one

Tweet media two

20

34

941

@daniela_witten

Daniela Witten

@daniela_witten

4 years

God grant me the confidence of a colleague who lectures me about how I need to “be aware that some people have childcare responsibilities” on a zoom call when I’m literally nursing a baby

7

20

925

@daniela_witten

Daniela Witten

@daniela_witten

2 years

academia is a war, and your only weapon is the length of your email signature

29

48

922

@daniela_witten

Daniela Witten

@daniela_witten

3 years

180 more pages but you can’t put a number on 🩸 😓 and 😭

@GueyeNono

Nono Gueye

3 years

Wow it is so much bigger than the 1st edition! @daniela_witten @HastieTrevor @robtibshirani #rstats #ISLR #ISLR2

Tweet media one

0

5

104

18

27

899

@daniela_witten

Daniela Witten

@daniela_witten

4 years

We’re celebrating the fact that international scholars are not going to be kicked out of the US in a month or two — a scenario that should absolutely never have been under discussion in the first place. So yeah, it’s been a rough 6 months. Let’s not confuse this with a victory.

7

110

871

@daniela_witten

Daniela Witten

@daniela_witten

3 years

A victory speech at the Chase Center is fine, but it's not as good as a victory speech at Chase Center Total Landscaping

11

51

865

@daniela_witten

Daniela Witten

@daniela_witten

4 years

shout-out to my first-ever reviewer #3 , who in 2006 wrote a review that read, in its entirety: "I'm pretty sure I saw similar work a few years ago from some group out of Australia" i will never forget u

24

41

861

@daniela_witten

Daniela Witten

@daniela_witten

4 years

Yes, it's just the name of a lecture. But we need to start somewhere. Statistics: we can, and should, do better. /fin

82

25

834

@daniela_witten

Daniela Witten

@daniela_witten

4 years

If you ever ask me how to do some statistical analysis and I tell you “look it up on Wikipedia” then please know I’m not being rude. I’m telling you the literal answer to your question, which is that I’d look it up on Wikipedia

15

70

822

@daniela_witten

Daniela Witten

@daniela_witten

4 years

As a statistician, I am very surprised that *both* D and M contracted it. It’s basically the intersection of two independent events since they never hang out

21

34

814

@daniela_witten

Daniela Witten

@daniela_witten

3 years

When I started as faculty in 2010, the first seminar I attended was in another department. I asked a question or two, and after the talk the speaker ran after me as I was leaving to ask if I wanted a paid research assistantship in their lab. That’s it. That’s the whole story

40

31

804

@daniela_witten

Daniela Witten

@daniela_witten

3 years

#zoomlife is crazy. I can give a keynote, hold office hours, and meet with a collaborator who lives 1,000 miles away all within a 2.5 hour window. No wonder we are all exhausted.

15

34

785

@daniela_witten

Daniela Witten

@daniela_witten

4 years

No matter how small the p-value for a regression of IQ onto shoe size is, that doesn’t mean that big feet cause smarts!! It just means that grown-ups tend to have bigger feet and higher IQs than kids. 3/

6

40

770

@daniela_witten

Daniela Witten

@daniela_witten

3 years

Women of academia: what career mistake did you make this week?

Not self confident enough

2353

Not a “team player”

485

Had a female mentor

1292

Other (write in comments)

788

126

122

768

@daniela_witten

Daniela Witten

@daniela_witten

3 years

One year late (due to COVID cancelling last year’s IMS World Congress) but still 🤩

Tweet media one

12

16

752

@daniela_witten

Daniela Witten

@daniela_witten

3 years

I just received an invitation to write a book chapter, which mentioned that I will be "compensated" with a hard copy of the book. Academia: I can't even

44

17

753

@daniela_witten

Daniela Witten

@daniela_witten

4 years

It is such an honor to receive the 2021 Breiman Award from @AmstatNews for "outstanding contributions to statistical learning, computational statistics, and their practical applications". 1/4

Tweet media one

43

30

740

@daniela_witten

Daniela Witten

@daniela_witten

4 years

academia is basically a whole bunch of people sitting in a room saying "yes, it's really too bad, but we didn't write the rules" when literally they wrote the rules

10

66

719

@daniela_witten

Daniela Witten

@daniela_witten

2 years

I have been a statistician for 16 years and I have yet to learn the Greek alphabet. Nice try, covid variants, but if years of typesetting math stats in LaTeX couldn’t teach it to me then you don’t stand a chance

19

23

719

@daniela_witten

Daniela Witten

@daniela_witten

4 years

But it's 2020, and there’s a need for us in academia to look hard in the mirror and see whether the things that we are doing are contributing to an environment that is welcoming for all people. I do not think that naming a lecture after a famous eugenicist is a good move! 17/

7

67

714

@daniela_witten

Daniela Witten

@daniela_witten

4 years

Life as a woman in academia: Male colleague #1 (MC1): I put a very rough draft in this google doc. Me: Thank you, MC1, for the great start! I just extensively edited the doc for the past 3 hrs. I think it's good to go. MC2: Thanks MC1 for single-handedly writing this document.

22

26

684

@daniela_witten

Daniela Witten

@daniela_witten

2 years

Breaking news from the CDC: starting now, the p-value is the probability that H0 is true

16

48

671

@daniela_witten

Daniela Witten

@daniela_witten

4 years

That’s it for today. Have a great weekend! 🌞⛱️🩴 13/13

52

25

665

@daniela_witten

Daniela Witten

@daniela_witten

4 years

@BryanDMartin_ You know I can see this right ?

6

11

668

@daniela_witten

Daniela Witten

@daniela_witten

5 years

Hi @reneegenn , we’ve never met but you seem remarkably well-adjusted & you have great hobbies. This book is light enough to stick in your carry on and it pairs well with basically all wine

@reneegenn

Renee Lynn (she/her)

5 years

Why yes I am the psycho who is bringing this book on her two week vacation to Italy. What better way to get started with machine learning? 🤓

Tweet media one

10

16

114

12

34

659

@daniela_witten

Daniela Witten

@daniela_witten

3 years

HUGE CONGRATS to @robtibshirani for winning the 2021 Founders of Statistics Prize. 🥇 🎉 🥳 👏 A well deserved recognition to a person who has written more groundbreaking papers than I can count.... and who continues to be a role model to me both professionally & personally 🙏

Tweet media one

7

56

649

@daniela_witten

Daniela Witten

@daniela_witten

4 years

THREE: SEEK AND YOU SHALL FIND. If you look at your data for long enough, you will find something interesting, even if only by chance! In principle, we know that we need to perform a correction for multiple testing if we conduct a bunch of tests. 11/

8

31

632

@daniela_witten

Daniela Witten

@daniela_witten

4 years

When I teach cross validation in my Intro to Statistical Learning course, I literally spend a class on “potential pitfalls of CV” and this is the main error I talk about. Happens all the time in published biology literature- not just for methylation data 1/

@jmschreiber91

Jacob Schreiber

4 years

The more papers I read for a review article I'm writing about ML pitfalls in genomics, the more my faith is shaken in the results from papers that apply machine learning to methylation arrays. A salty thread. 1/

21

247

897

11

147

634

@daniela_witten

Daniela Witten

@daniela_witten

4 years

I took my 1st CS course at age 18. I almost dropped it 3 days in b/c I couldn’t install the software. My friend had to help me install it. We are married now. Sometimes I make him install the updates on my Mac, for old times sake. True story.

@KLdivergence

Kristian Lum

4 years

To add some interesting color to this story (bc who doesn't love a good story arc?), it's crazy to me that I'm on the faculty of a CIS department. My first semester in college I started an intro CS course. I was super pumped about the topic. I dropped it two weeks in.

2

3

65

17

24

633

@daniela_witten

Daniela Witten

@daniela_witten

2 years

The plumber who we called to fix our leaking toilet told us that whoever installed it 8 years ago must have been a complete hack. Friends, he’s the person who installed it 8 years ago

22

13

624

@daniela_witten

Daniela Witten

@daniela_witten

4 years

So, unless you can design your study to uncover causation (very hard to do in most practical settings — the field of causal inference is devoted to understanding the settings in which it is possible), the best you can do is to discover correlations. Sad but true. 4/

5

30

605

@daniela_witten

Daniela Witten

@daniela_witten

2 years

This is correct

@xkcdComic

XKCD Comic

2 years

Confounding Variables

Tweet media one

27

1K

6K

10

64

601

@daniela_witten

Daniela Witten

@daniela_witten

1 year

Everybody please stop with the “this is the end of twitter” posts. I have 3 kids and a full time job. I don’t have time to find another hobby

9

25

594

@daniela_witten

Daniela Witten

@daniela_witten

1 year

I don’t expect ChatGPT to change academia because writing long-winded but eloquent text that has a non-zero probability of being hot garbage is not a core academic competency.

31

20

592

@daniela_witten

Daniela Witten

@daniela_witten

2 years

Oops, I mean to say, she prefers R to python so she’s an 11 (R indexes from 1)

@daniela_witten

Daniela Witten

@daniela_witten

2 years

She’s a 10 but she prefers R to python, so, yeah, she’s still a 10

26

178

3K

7

32

589

@daniela_witten

Daniela Witten

@daniela_witten

6 years

This would be nice, since right now I’m a full time professor with a very unfulfilling part time gig organizing my travel reimbursement receipts.

@davegray

Dave Gray

6 years

Google discovered that giving every employee a credit card and trusting them to follow policy on expenses, even with some bad actors, was cheaper than the enforcement and structures most companies use. How much do you waste due to low trust? Can you do the math for your company?

222

4K

11K

16

72

581

@daniela_witten

Daniela Witten

@daniela_witten

5 years

One of my students wrote in my course evaluation that they "really love the professor's nerd humor". Thanks and I appreciate the sentiment, but at the same time it stings a bit because that was just my regular humor 🤓🤦‍♀️😭

12

10

566

@daniela_witten

Daniela Witten

@daniela_witten

4 years

This made it into my twitter bio

Tweet media one

@daniela_witten

Daniela Witten

@daniela_witten

4 years

excerpt from an e-mail I just received. honest to god, I have (to my knowledge) never interacted with this person in my life. they really need a hobby that does not involve e-mailing me.

Tweet media one

43

5

246

20

6

569

@daniela_witten

Daniela Witten

@daniela_witten

4 years

But in practice, what if we decide what test(s) to conduct AFTER we look at data? Our p-value will be misleadingly small because we peeked at the data. Pre-specifying our analysis plan in advance keeps us honest… but in reality, it’s hard to do!!! 12/

5

34

558

@daniela_witten

Daniela Witten

@daniela_witten

4 years

Checking in for a flight from Canada back to Seattle. The woman at the @AlaskaAir checkin desk asked me if I’ve been to China or Iran in the past 14 days. “No”, I replied, “but I have been to Seattle.” Friends, she did not even crack a smile. #CoronaVirusSeattle #coronavirus

11

49

549

@daniela_witten

Daniela Witten

@daniela_witten

3 years

It's here!

@daniela_witten

Daniela Witten

@daniela_witten

3 years

The second edition of ISLR is due out by end of summer! If you are an **instructor** in need of a pre-print to prepare for AY 2021-2022, then send an e-mail to hello @statlearning .com

Tweet media one

24

262

2K

5

70

553

@daniela_witten

Daniela Witten

@daniela_witten

2 years

Any meeting that lasts more than 60 minutes and involves more than 6 people should have snacks, there are no counterexamples, thanks for coming to my Ted talk

24

31

549

@daniela_witten

Daniela Witten

@daniela_witten

4 years

Often, people focus on that tiny p-value, and the fact that the effect is of **literally no practical relevance** is totally lost. 8/

5

25

546

@daniela_witten

Daniela Witten

@daniela_witten

3 years

Plz, plz, plz, HELP ME (& other senior faculty) HELP YOU. Make a good website --- easy to find, organized, & up-to-date. I'm eager to support & promote junior researchers. But if I can't find key info about you from a quick Google search then I'll move on to the next person.

21

58

541

@daniela_witten

Daniela Witten

@daniela_witten

4 years

If you’re testing whether mean=0 and actually the truth is that mean=0.000000001, and if you have a large enough sample size, then YOU WILL GET A TINY P-VALUE. 6/

8

30

535